I have attached my comments FOLLOWING the 3 messages posted by Julie, Tim and Bob re AP Stats Software. =================================== JULIE REULBACH WRITES ===========
From email@example.comTue Apr 9 10:07:24 1996 Date: Mon, 8 Apr 1996 17:33:02 -0400 (EDT) From: Todd and Julie Reulbach <firstname.lastname@example.org> To: email@example.com Subject: AP Stats software
Hello! My school system is going to be starting the AP Stats class and will have great access to computers. I was wondering if anyone could recommend computer software for the course. The AP Stats info book had a listing of many, but I don't know what is good (or NOT good). If anyone has any info on either one, please let me know!!
Thanks, Julie Reulbach
========================================================= TIM BROWN WRITES ======
From firstname.lastname@example.orgTue Apr 9 10:06:17 1996 Date: Tue, 09 Apr 1996 08:09:21 -0400 From: Timothy Brown <email@example.com> To: firstname.lastname@example.org, email@example.com Subject: AP Stats software -Reply
Welcome, Julie. There's been a lot of discussion about computer software on this list over the past few months, and if my memory serves me the most commonly mentioned packages (for the Mac, anyway) are MINITAB and DATA DESK. My preference is MINITAB, but I'm not an expert, and you'll get a score of other rec's.
I'm responding to you message because I'd like to hear from the group about exactly how important people feel computer software *is*. I am trying to teach the AP syllabus for the first time this year, and I find that it takes a conscious effort to involve the computer, because *everything* can be done with the TI-82 and some freeware that is readily available at NCTM conferences (stuff that I suspect will be built-in on the TI-83). Why do we need computers? It feels artificial to employ a sledgehammer to drive a thumbtack. Especially when every student can have the calculator in their hands.
Tim Brown Lawrenceville School Lawrenceville, NJ 08648 609-895-2050
============== BOB HAYDEN WRITES -- ===============
From firstname.lastname@example.orgTue Apr 9 10:04:38 1996 Date: Tue, 9 Apr 1996 09:56:21 -0400 (EDT) From: Bob Hayden <email@example.com> To: firstname.lastname@example.org Subject: calculators
> From: hayden (Bob Hayden) > Subject: AP Stats software -Reply (fwd) > To: APstats > Date: Tue, 9 Apr 1996 09:53:10 -0400 (EDT) > > I think the calculator makes it too easy to be content with number > crunching and to avoid real data analysis. If you just want to get > the mean of five numbers and that's the end of it, the calculator is > fine. If you want to work all semester with several large, real data > sets, you quickly come up against the limitations of the calculator. > Just one moderate size data set with a few variables will exhaust its > storage capacity. If you want to then look at another data set and > later return to the first, you have to constantly retype the data or > connect to someone who still has it on their calculator or download it > from a commputer (in which case, why not use the computer to analyze > the data?). Now, suppose you want to compare the normal probability > plot for this week's data to the same plot for last weeks data? > > On the output side, I felt like we made a big step BACKWARDS at my > school when we moved from paper printing terminals to CRTs. I wanted > my students to be able to compare several different displays of the > same data. The displays will not fit on a single CRT screen, to say > nothing of a calculator screen. > > I think the calculator is an impediment to thoughtful data analysis. > If you did all your dispalys by hand, then at least you could go back > and look at them next week. > -- _ | | Robert W. Hayden | | Department of Mathematics / | Plymouth State College | | Plymouth, New Hampshire 03264 USA | * | Rural Route 1, Box 10 / | Ashland, NH 03217-9702 | ) (603) 968-9914 (home) L_____/ email@example.com fax (603) 535-2943 (work)
================ JOE WARD COMMENTS ---------------- I agree with Bob Hayden's comments and I will add a few thoughts related to the Subject.
SOME FACTS --
Interest in statistics as the secondary level has been heightened with the announcement that on May 6, 1997, the Advanced Placement Examination in Statistics will be given for the first time by The College Board. Of particular significance is the role of technology in AP Statistics.
As indicated in the Preliminary Edition of the Advanced Placement Course Description for Statistics, "The fundamental tool of data analysis is the computer." (p. 9). "Because the computer is central to what statisticians do, we believe it is essential for teaching the AP Statistics course." (p. 10). "While it would be ideal for students to have access to a computer during the AP Statistics Examination, this is currently unrealistic. Thus, graphing calculators will be required and computer output will be provided a necessary. Familiarity with generic computer output may be necessary to answer some questions on the examination." (p. 11). A copy of this announcement can be obtained from the Educational Testing Service (ETS) or Regional Offices of the College Board. The email addresses are shown below. ========= BEGIN JOE WARD OPINIONS (POSSIBLY SHARED BY MANY OTHERS) --------- It is apparent that it will be necessary for teachers of statistics to choose an appropriate statistical software package to facilitate data analysis. Several factors that seem significant in the selection of the statistical software system are:
1. The teacher's personal confidence in using the selected software.
2. Compatibility of the software used for the examples in the selected text with the chosen software. In selecting a statistics text, it would be desirable to have a text that uses the selected software for the data analysis examples.
3. The desirability of using "Student Versions" of larger statistical packages. This consideration would allow for the acquisition of multiple copies of the "Student Version" and perhaps one copy of the larger system. The knowledge obtained with the Student Version would allow students to easily use the larger system.
4. The cost of the software. ============== BEGIN JOE WARD OPINIONS (UNLIKELY SHARED BY OTHERS)
The AP-Statistics Exam certainly is an important step toward stimulating interest in statistics the high schools. I whole-heartedly support the AP-Statistics Development Committee's specifications for the AP Statistics Course. I feel strongly that their statements "The fundamental tool of data analysis is the computer." (p. 9)... "Because the computer is central to what statisticians do, we believe it is essential for teaching the AP Statistics course." (p. 10). should be printed in large, bold letters.
My personal opinion is that the Development Committee has done a good job in providing a course description that possibly will satisfy a large number of colleges who will give college credit for statistics. However, it is my observation that a majority of the college one-semester non-calculus based statistics courses are focusing on what I would call "the pre-computer ways of statistics instruction". There seems to be a reluctance to provide students with the POWER THAT THEY DESERVE. It may be that us "old folks" think that if we suffered with algorithms designed for pre-computer days, that our students should not be allowed to skip the suffering. For example, some teachers and text books would have their students believe that in a "two-factor ANOVA they must have equal or proportional cell frequencies with NO MISSINGS CELLS" and that is why they must do the computing using the pre-computer algorithms. Hopefully the secondary teachers will be able to take the lead in giving THEIR students the power that they deserve.
Back to the Calculators --
The graphing calculator does lots of great things such as easy plotting of functions. But the main advantage is the low cost.
For at least five years I have talked with almost every calculator rep about the simple requirement to allow the user to have the option to NOT INCLUDE Y-INTERCEPT IN A SIMPLE REGRESSION EQUATION.
Now, the user is REQUIRED to use the linear model:
Y = a + b*X
(where a is the Y-intercept and b is the "least-squares" coefficient for X. It is called the "slope of the line in the X-Y plane" when the researcher believes that the "amount of change in Y per unit change in X is constant". However, the researcher may wish to let the predictor X contain ANY VALUES such as t^2, t^3,..., cos(t), log(t) , etc. For example, D = k1*T^2)
So if the user really believes that the function should pass through the origin -- it's not easy.
AT THE PRESENT TIME, TI-83 DOES NOT ALLOW THE USER TO REMOVE THE CONSTANT!
And furthermore, most interesting problems are of the form:
Y = a + b1*X1 + b2*X2 + b3*X3 + ...
and in some cases the model may be of the form
Y = c1*X1 + c2*X2 + c3*X3 + ... (NO CONSTANT)
Of course, these problems can be solved using the matrix operations -- which may be outside the learning objectives for the students.